Claude Code plugin: write a full-length reference or instructional book with teams of subagents. Definition, planning, research, parallel chapter drafting, review, graphics and publication — organised so no single agent is ever given a task large enough to degrade. Authorship only; print geometry hands off to kdp-publishing.
1 on GitHubby Daniel Rosehillv1.0.0MITupdated 2 months ago
Not in a marketplace yet, so it loads for one session with --plugin-dir.
Run the definition stage of a book project — record the brief verbatim, interview for the gaps, and settle the spec. Use at the start of a book, before any planning or outlining.
book-drafting
Dispatch chapter-writing subagents in parallel batches to draft the book, and handle research where a chapter's sources are not yet gathered. Use once briefs exist. Enforces the batching rule that keeps parallel writers from colliding.
book-graphics
Produce the book's figures — establish the graphics style guide, spec each figure, and generate them. Use after chapters are drafted, in parallel with review.
book-planning
Run the planning stage — write the style guide, design the structure, and produce one brief per chapter. Use after the spec is settled and before any drafting.
book-publication
Assemble the drafted chapters into a manuscript, typeset the body to the chosen profile, and hand off to print geometry. Use as the final stage, after review and graphics.
book-review
Run the review team over drafted chapters — proofreading, voice-drift audit and continuity audit. Use after drafting and before publication. Skippable by explicit user choice; the mechanical gates are not.
create-workspace
Create a new book project workspace from the Claude-Book-Workspace-Template repository, or scaffold one locally. Use when starting a new book, or when the user asks to set up a book project.
Slash commands (12)
/book-writing:book-assemble
Assemble the manuscript and hand off to print geometry
/book-writing:book-brief
Record the user's brief for the book exactly as given, asking nothing
/book-writing:book-draft
Draft chapters with parallel writing subagents
/book-writing:book-graphics
Establish the visual language and produce the book's figures
/book-writing:book-interview
Interview the user for the gaps the writing team will need — and only those
/book-writing:book-new
Start a new book project — create the workspace repository from the template
/book-writing:book-outline
Design the structure and write one brief per chapter
/book-writing:book-research
Gather the source material a chapter needs
/book-writing:book-review
Run the review team — proofreading, voice drift, continuity
/book-writing:book-spec
Settle the brief and interview into the book spec
/book-writing:book-status
Show what is drafted, what is not, and what is blocking production
/book-writing:book-style
Write the style guide — the binding voice reference every agent reads
Subagents (12)
brief-recorder
Records the user's brief for a book exactly as given, without asking questions or interpreting it. Use as the very first step of a book project, before any interview or planning. Writes spec/brief.md and stops.
chapter-researcher
Gathers and records the source material one chapter needs, working from that chapter's brief. Use before drafting a chapter whose sources are not already at hand. Writes research/notes/*.md and rows in research/sources.md. Does not draft.
chapter-writer
Drafts exactly one chapter from its brief, the style guide and its research notes. Never reads other chapters. Use for every chapter of the book, dispatched in parallel batches. Writes manuscript/<chapter>.md and its rows in the continuity ledger.
continuity-auditor
Checks the book for repetition, contradiction and unpaid promises across chapters, using lint output and the continuity ledger rather than full drafts. Use once most chapters are drafted, and again before production. Writes review/reports/continuity-*.md; never edits prose.
figure-maker
Produces one figure from its spec, obeying the graphics style guide. Use per figure after specs exist. Writes graphics/diagrams/ sources and graphics/assets/ output.
graphics-director
Owns the book's visual language — writes the graphics style guide and one spec per figure, deciding what each figure must show. Use after chapters are drafted and before any figure is made. Writes guides/graphics-style-guide.md and graphics/specs/*.md.
outline-architect
Designs the book's structure and writes one brief per chapter — the contracts that make parallel drafting possible. Use after the spec and style guide are settled. Writes outline/outline.md, outline/briefs/*.md and the chapter entries in book.toml.
proofreader
Reads one chapter for mechanical correctness — grammar, spelling, punctuation, terminology consistency and style-guide compliance. Use per chapter after drafting. Writes findings to review/reports/; never edits prose.
spec-interviewer
Interviews the user to fill only the gaps the book-writing team will actually need — the questions whose answers change the outline, the voice or the scope. Use after the brief has been recorded and before the spec is written. Writes spec/interview.md.
spec-writer
Settles the recorded brief and the interview into the book spec — what the book is, who it is for, what it must not become, and its shape. Use after intake and before any planning. Writes spec/spec.md.
voice-auditor
Checks the book for tone-of-voice drift against the style guide by sampling across chapters. Use after several chapters are drafted, and again before production. Writes review/reports/voice-*.md; never edits prose.
voice-designer
Writes the book's style guide — the binding voice reference every writing and review agent reads. Use after the spec is settled and before the outline. Writes guides/style-guide.md.
AgentDB (external: maintained in ruvnet/agentdb, pinned to commit 27678c3, plugin 3.6.0): agentdb's own standalone MCP server pinned to [email protected] (vector memory with get/list/update/delete for records, patterns and skills, hybrid keyword + vector episode search and an opt-in local cross-encoder, an opt-in aidefence injection guard on memory writes, reflexion, skills, outcome feedback, capabilities report), a memory-specialist agent, 3 skills, /agentdb-status, and the agentdb mod guard that keeps secrets out of memory and refuses delete/reset unless allowed. Complements ruflo-agentdb, which reaches AgentDB through ruflo's server
Provision direct cloud access (Vertex AI, Bedrock, or LLM gateway) for the Claude Microsoft 365 add-in. Generates the customized manifest, walks through Azure admin consent, and writes per-user config via Graph extension attributes.
Build, launch, grade, and schedule Claude Managed Agents (CMA) in your own Anthropic account — a plugin re-implementation of Anthropic's launch-your-agent reference skill (Apache-2.0; independent, not a fork). Every session starts with a goal (./my-agent/goal.json, opt-in SessionStart hook, driven by /cs-goal); loop_compiler.py compiles it into a bounded grade->iterate loop (max_iterations 1..20), a recurring cron scheduled-deployment loop, or a single-pass interview->stage->launch workflow. 6 skills: orchestrator (context: fork goal router) + interview + stage-launch + grade-iterate + run-without-you + wrap-up. 18 stdlib-only deterministic scaffolders (no network/API calls; live launches emitted as BYOK curl that never prints the key), 4 agents, 8 commands, hooks, 5 references, 4 assets. Validators enforce CMA limits. Distinct from engineering/agent-harness (generic domain loop) and write-a-skill (authors Claude Code skills, not CMAs).
6 skills · 8 commands · 4 agents · hooks
alirezarezvani/claude-skills 29,725
Maintain book-writing?
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.